Even when acute myelogenous leukemia (AML) initially responds to standard chemotherapeutic regimens, it has a high relapse rate attributable to persistence of minimal residual disease (MRD). Both experimental and clinical data support a contribution from the immune response in the elimination of MRD in AMI. Tumor associated antigens and minor histocompatibility antigens are potential targets for immunotherapy following chemotherapy or hematopoietic stem cell (HSC) transplantation, respectively. A major requirement for an effective immune response to experimental tumors is the generation of tumor-specific CD4+ T lymphocytes. It is likely that recognition of MHC class ll-restricted epitopes by the immune system will therefore be essential for immunotherapy of naturally occurring malignancies, including AML. As yet, few class ll-restricted tumor antigens or minor histocompatibility antigens have been identified in immune responses against leukemic blasts. To improve the prospects for successful AML immunotherapy I have adapted a molecular screening approach to identify potential leukemia tumor antigens. I am using a vector encoding the human invariant chain to enhance the class II presentation of peptides from tumor cDNA libraries. I have already successfully used this targeted gene expression method to identify a class II epitope of the LAGE-1 tumor antigen in melanoma, and described the antigen-specificity of regulatory T cells in this cancer (Immunity, 2004). To extend this technique to human AML I have developed methods for the recovery of cancer-specific T cells from human bone marrow samples and for the expansion of leukemic blasts in NOD-SCID mice. I propose to use these cellular and molecular techniques on blood or bone marrow samples obtained from patients with AML at diagnosis, following autologous tumor vaccination, or after HSC transplantation to: 1) identify novel class I and class ll-restricted tumor associated antigens responsible for cellular immune responses against leukemic blasts in AML patients, 2) characterize the peptide epitopes of these tumor antigens, and 3) follow the development of anti-cancer immune responses to these epitopes following HSC transplantation or tumor vaccination. I anticipate that this translational research could lead directly to clinical trials evaluating antigen specific active (vaccines) or passive (adoptive transfer) immunotherapy for AML.